JPS6171743A - End delimiter provision system in loop network - Google Patents

Abstract

PURPOSE:To restore a system quickly to the normal state even if a signal not eliminated remains on a transmission line by forming an end delimiter of a transmission frame of a trigger station different from a form of an end delimiter given to the final transmission frame of each node station. CONSTITUTION:End delimiters TED, NED are given respectively to a trigger station TS and a node station NS. Suppose that a transmission error is caused on the way of a packet b1 and a pseudo end delimiter END is produced as shown in a black mark (x). The NS22 transmitting the packet b1 recognize in error the NED as an MED of the packet b1 and completes the elimination control. Thus the rest of the packet b1 is fed to an NS23. The NS23 eliminates a part of the packet b1 but passes the own transmission packet c1 toward the lower stream. The TS20 is different from each NS, the elimination control is not finished by the passing of the NED of the said packet c1, but finished by the detection of the end TED of the polling signal p2. Thus, the own signal p2 and the c1 not eliminated by the NS23 are eliminated in the TS20 and the system restores the normal state.

Description

Detailed Description of the Invention (Field of Industrial Application) The present invention provides a method for adding an end delimiter to a signal frame in a loop network to reduce the effect of code errors on a transmission path even if they occur. Regarding.

(Prior art) In addition to the integration of voice communication and data communication, as a communication method to realize the integration of wideband services such as video communication and high-definition images, for example, 1982 Institute of Electronics and Communication Engineers Exchange Study Group Material 8E83-107 There is a loop-like network described in ``A Study of High-Speed Synchronous Packet Route 1''.

This loop-like network constitutes a network as shown in FIG. In the figure, trigger station controller T820 transmits a polling signal onto transmission line 24, and node stations N521, N522, . N523
Performs polling control for. Also, the node stage, N821. N522 and N523 transmit and receive packets via the transmission path 24 in accordance with the polling control. These stations are connected in a loop through a transmission line 24 as shown in the figure. (In this specification, the node station is abbreviated as NS, and TS and NS are collectively referred to as stations.) In this loose network, as shown in FIG. , Ta2O transmits a polling signal (pi or p2), and each NS monitors the polling signal passing on the transmission path 24, prepares to transmit a packet, and transmits the packet after acquiring the transmission right. . For example, in the example shown in Figure 6, N52
1 receives the polling signal p1 transmitted by Ta2O, acquires the transmission right after passing the polling signal p1, and transmits (b)al. After receiving the polling signal p1, N522 acquires the transmission right after passing the bucket)al and transmits the packet b1. Similarly, N523 also sends bucket) cl.

When the polling signal p1 transmitted by itself returns after going around the transmission path 24, Ta2O receives the polling signal p1.
1 is removed from the transmission path 24. In the figure, the signals removed at each station are marked with *. ), and further N5
After passing the packets al, bl, cl sent by 21, N822, and N823, they acquire the transmission right, transmit the next polling signal p2, and send the bucket to N821)a
Start sending 2. In addition, when each N3 transmits a packet that it has sent after going around the transmission path 24 and returns, the transmission path 2
Remove from 4. In this way, the transmitted signal is removed at the transmission source station after making one circuit around the transmission path 24.

(Problems with the prior art) In the loop network, each station is
When starting signal transmission, the transmission line 24 is temporarily disconnected, the signal within its own station is transmitted, and this signal is transferred to the transmission line 24.
The signal is removed from the transmission line 24, and then the transmission line 24 is connected again. In this way, each station controls the opening and closing of the transmission line 24 during transmission. Therefore, stage 1 is affected by code error υ on the transmission line 24.
(If the transmission line 24 is opened or closed by the operator, the transmitted packet may be deleted at a certain station before reaching the destination station), or even if the packet has completed one round on the transmission line 24. Regardless of the source station, a failure such as not being removed may occur, causing an abnormality in the access method of each station. For example, in the example of Figure 7, the packet sent by N821)
If an end delimiter indicating the end of the packet with a code error appears in the middle of the packet, as shown by the heat mark, NS2.1 that sent the bucket
When the end delimiter caused by the code error is detected when the end delimiter generated by the code error is detected, the removal of the bucket) al is completed.

Therefore, as shown by the f+ line in FIG. 7, a part of the latter half of the packet a1Ω is left without being removed.

On the other hand, N522 identifies this remaining signal as the bucket) bl transmitted by itself, disconnects the transmission line 24 until it detects the end delimiter of the signal shown by diagonal lines, and returns to the over-removal state. Therefore, even if the bucket )bl sent by N522 goes around the transmission path 24 and returns to the source station N522, it remains unremoved at the same stage and remains υ, N52
It is removed in 3. Similarly, the packet c1 transmitted by N823 is removed at Ta2O, the polling signal p2 transmitted by Ta2O is removed at N521, and so on. From then on, each source station transmits a signal different from the packet it transmitted. Abnormal conditions that result in the removal of the material will continue.

(Object of the Invention) The object of the present invention is to provide a method for adding an end delimiter to a signal frame that can quickly restore the normal state even if a signal remains on the transmission path that is not removed due to a code error on the transmission path. It's about doing.

(Structure of the Invention) According to the second method of assigning an end delimiter of the present invention, the end delimiter given to the last signal frame sent out from among the signal frames sent out by the trigger station is applied to the end delimiter given to the signal frame sent out by each node stage. It is possible to make the end delimiter different from the end delimiter given to the last signal frame to be sent.

(Detailed Description of Configuration) The assignment of an end delimiter to a loop network according to the present invention and the access control based thereon will be described in detail with reference to the drawings. Figure 2(a).

(bl shows the frame structure of the polling signal transmitted by the Tl 920 and the structure of the packet signal transmitted by each NS described above. These signals include a header indicating the originating station address, terminating station address, frame control information, etc. It consists of a message part (HDR) and a message part (INFO) that indicates the communication content, and delimiters are added before and after this to indicate the start and end of the signal.However, although the start delimiter (SD) is common, it indicates the end. The delimiter is different p, giving the ending delimiter TED for the signal frame transmitted by Ta2O and giving the ending delimiter NED for the signal frame transmitted by NS 0 T S 20. All stations including each NS ,
The first signal received from the upstream side of the transmission line 24 after the transmission is completed is always the own signal that has gone around the transmission line 24, as shown in FIG.
A polling signal pi is received at N521.22 and 23 are received packets al, bl and cl, respectively. Therefore, Ta2O detects the end delimiter TED, and each NS recognizes the cycle of the packets it has sent by detecting the end delimiter TED, and ends the removal control of the transmitted packets as in the prior art.

FIG. 1 is a diagram showing an embodiment of the trigger station T820 and each node stage generator NS according to the present invention.
Trigger stage TS20 in the same figure (a) is switch 1.
．． TED detection section 2. Transmission right acquisition unit 3. Access control section 4
．． It consists of a transmission register 5. The transmission right acquisition unit 3 determines the end delimiter (TED) of the signal frame passing on the transmission path 24.
orNED), and if immediately after that is in an idle state, the access control unit 4 is notified of acquisition of the transmission right. Based on this K, the access control unit 4 switches the switch 1 from the transmission path 24 side to the transmission register 5 side. The transmission register 5 is supplied with polling signals HDR and INFO to be sent, as well as a start delimiter SO and an end delimiter TBD, and this polling signal is sent to the transmission line 24 via the switch 1.On the other hand, the TED detection unit 2 teeth.

The access control unit 4 monitors the signal supplied from the upstream side of the transmission path 24, and when the end delimiter TED is detected, the access control unit 4
to notify. Based on this, the access control unit 4 connects the switch 1 to the transmission line 24 side and returns it to the passing state. By this control, the polling signal that has made one round on the υ transmission line 24 is removed, and the packet signal that follows it is bypassed and supplied to the downstream node station.

Each node station NS in FIG. 1(b) includes a switch 31, an NgD detection section 32, a transmission right acquisition section 33, an access control section 34, and a transmission register 35. This configuration is the same as Ta2O except that the NED detection unit detects the end delimiter NED of the signal frame received from the transmission path 24, and the transmission register 35 sends out a packet signal with the end delimiter NBD.◎ This implementation The operation of the example will be explained using FIG.

Ta2O sends out a polling signal p1 to which an end delimiter TED is added. N521y 22, 23 detects the idle state of the transmission line 24 after passing pl, and outputs packets al and 24 with end delimiters NED added, respectively.
Send bl and cl, respectively. After the polling signal goes around the transmission path 24, it is removed at the TS 20 (indicated by *). '1' When the S 20 detects the end delimiter TED, it ends the removal control and passes the signals below the bucket)at that are subsequently received.

And under normal conditions, bucket) al,? +1°cl
are removed from the transmission line 24 at N521, 22, and 23, respectively. However, now suppose that a pseudo-end delimiter NED, indicated by a heat mark x, occurs due to a transmission error in the middle of packet b1. The N822 that sent ble (bucket) mistakenly recognized this pseudo-end delimiter NED as the NED of packet b1,
End removal control.

Therefore, the rest of the packet bi shown in the shaded area is not removed and is supplied to N523. Since N523 ends the removal control based on the end delimiter NED that is detected first, 0TS20, which removes part of the packet and )bl but passes its own sending bucket )cl downstream, differs from each NS. , by the detection of TED), the removal control is terminated.
The removal control is not stopped when the bucket (cl) that could not be removed at N523 passes the end delimiter NED, but is ended by the detection of the end delimiters TF, D of the polling signal p2).

Therefore, in Ta2O, cl that N523 could not remove from the polling signal p2 sent by itself is removed,
Return to normal state.

On the other hand, a case will be explained in which a pseudo end delimiter TED occurs in the middle of a packet signal due to a transmission error. Now, assume that a pseudo delimiter TBD, indicated by an unmarked Y, occurs in the bucket )al. Since N521 does not terminate removal control based on TED, access control is not affected even if such an error occurs.As explained above, even if a transmission error occurs in a packet on the transmission path, the next polling Does not affect control.

Although the case where each station transmits only one signal frame has been described, the present invention can also be applied to a case where a plurality of signal frames are transmitted. In this case, as described in Patent Application No. 1984-084581 "Transmission Control Method in R-Loop Network", the end delimiter of packets other than the last packet to be sent includes an end delimiter IED indicating that the packet is an intermediate packet. Add. Fourth
Figure (a).

(bl shows how to give the end delimiter of each signal frame when the trigger station TS20 and the node station NS concatenate and transmit three packets.

(Effects of the Invention) If the end delimiter adding method of the present invention is used, even if a false end delimiter is generated in a packet on the transmission path due to a code error and an abnormal state occurs, the normal state can be restored at the trigger station. Abnormal conditions will not continue for a long time.

[Brief explanation of drawings]

Fig. 1 shows the configuration of the trigger stage and node station, Fig. 2 shows the structure of the signal frame, Fig. 3 shows the flow of the signal frame on the loop-shaped transmission path, and Fig. 4 shows the structure of the signal frame. The figure shows how to give an end delimiter when transmitting multiple signal frames, FIG. 5 shows the configuration of a loop network, and FIGS. 0 which is a diagram showing the flow
In the figure, 1.31 is a switch, 2.32 is an end delimiter detection unit, 3 and 33 are transmission right acquisition units, 4.34 is an access control unit, 5.35 is a transmission register, 20 is a trigger stage register, and 22 , 23° and 24 indicate the node stage numbers. Kyou 1 Figure (a) 3 (b) 3.3 Tei Z Ki 1F! PM--Middle S

Claims (1)

[Claims] A termination delimiter provision method in a loop network consisting of one trigger station and a plurality of node stations, each of which is connected to a loop-shaped transmission path, and each node station controls transmission based on a polling signal from the trigger station. The end delimiter given to the last signal frame sent out by the trigger station is different from the end delimiter given to the last signal frame sent out by each node station. A method for adding an end delimiter to a loop network.